The National Cancer Institute has made a substantial investment in new technology platforms for cancer proteomics, especially through the Mouse Models of Human Cancers and the Clinical Proteomic Technologies for Cancer. The proteome is critical to understanding functional genomics and systems biology of cancers and to discovery and validation of biomarker candidates and molecular targets for therapy and prevention. Sophisticated analysis of proteomes requires advanced informatics to deal with the complexity of specimens, the extreme dynamic range of protein concentrations, post-translational modifications, alternative splice isoforms, responses to all sorts of perturbations, and differences in databases. The current scarcity of trained scientists in this subdiscipline of bioinformatics is the focus of this proposed T32 training grant in Advanced Proteome Informatics of Cancer at the University of Michigan. We are building upon our successful experience with the Bioinformatics Graduate Program, based in the university-wide Center for Computational Medicine and Bioinformatics (CCMB). We now have 31 PhD and 6 M.S. students, plus 16 PhD and 14 M.S. graduates. Our faculty and students are in the leadership of Human Proteome Organization (HUPO) initiatives, development and global deployment of the Tranche distributed file-sharing system and the ProteomExchange, and creation of new algorithms for proteome informatics. We have a robust community of cancer researchers, bioinformaticians, statisticians, chemists, and software engineers focused on major challenges in proteome data analysis. Trainees will come from diverse backgrounds and will receive training in cancer biology, bioinformatics, and computer science through courses, seminars, journal club, and annual retreats. We have made substantial progress building strong relationships with sources of applicants from disadvantaged minority backgrounds. Establishment of an NCI training program in Advanced Proteome Informatics of Cancer at the U of M will provide a new pool of scientists well-equipped for independent careers in this field, enhance faculty research, and support NCI goals. RELEVANCE: Very large, complex datasets are being generated by cancer researchers using new proteomics technologies that allow them to study thousands of proteins simultaneously. This avalanche of data requires scientists well-trained in the specialized and multidisciplinary field of Proteome Informatics. Insufficient numbers of scientists are being trained in this rapidly growing field and this T32 proposes to address this issue.